SHUNT POWER ACTIVE FILTERS FOR DC DRIVE LOADS

Abstract

As industry power demand becomes increasingly sensitive, power quality distortion becomes a critical issue. The recent increase in nonlinear loads drawing non- sinusoidal currents has seen the introduction to manage the clean delivery of power. In this thesis, a three-phase shunt active power filter (APF) using artificial neural network technique is proposed to mitigate harmonics, to compensate reactive power, to improve power factor and to remedy system unbalance of DC drive loads. This thesis proposes the use of artificial neural network (ANN) technique for the shunt active filter. The extraction circuit uses ANN algorithm to compute the harmonic currents and reactive power for the nonlinear load. These two . signals are used as the reference signal for the hysteresis control of a thre•e­ phase IGBT voltage source inverter (VSI). With the use of this ANN extraction circuit, the shunt active filter can be made adaptive to variations in nonlinear load currents. It can also compensate for unbalanced nonlinear load currents and correct the power factor of the supply side near to unity. Each module of the proposed adaptive shunt active filter is discussed in detail. Computer simulations are carried out on three phase three wire APF's to verify the design conceptofthis adaptive shunt active filter.